5c1891ec9f
In the etcd container, the etcd data directory is always /var/lib/etcd. Reverting to this value, since `etcd_data_dir` on the host maps to `/var/lib/etcd` in the container.
7 years ago
61b74f9a5b
updated to direct control over ipip
7 years ago
7332679678
Give more control over IPIP, but with same default behaviour
7 years ago
91dff61008
Fixed helm bash complete
7 years ago
8203383c03
rename almost all mentions of kargo
7 years ago
fff0aec720
add configurable parameter for etcd_auto_compaction_retention
7 years ago
266ca9318d
Use the kube_apiserver_insecure_port variable instead of static 8080
7 years ago
db3e8edacd
Fixing up vault variables
7 years ago
6e41634295
Set default value for kube_hyperkube_image_repo
Fixes #1334
7 years ago
01c0ab4f06
check if cloud_provider is defined
7 years ago
56b86bbfca
inventory hostname for cordoning/uncordoning
7 years ago
7e2aafcc76
add direct path for cert in AWS with RHEL family
7 years ago
af55e179c7
adding --skip-exists flag for peer_with_router
7 years ago
d47fce6ce7
upgrade k8s version to 1.6.4
7 years ago
7ae5785447
Removed the other unused handler
With live-restore: true, we don't need a special docker restart
7 years ago
ef8d3f684f
Remove unused handler
Previous patch removed the step that sets live-restore
back to false, so don't try to notify that handler any more
7 years ago
cc6e3d14ce
Add host-based kubelet deployment
Kubelet gets copied from hyperkube container and run locally.
7 years ago
6f67367b57
Leave 'live-restore' false
Leave live-restore false to updates always pick
up new network configuration
7 years ago
9ee0600a7f
Update handler names and explanation
7 years ago
30cc7c847e
Reconfigure docker restart behavior on atomic
Before restarting docker, instruct it to kill running
containers when it restarts.
Needs a second docker restart after we restore the original
behavior, otherwise the next time docker is restarted by
an operator, it will unexpectedly bring down all running
containers.
7 years ago
a5bb24b886
Fix docker restart in atomic
In atomic, containers are left running when docker is restarted.
When docker is restarted after the flannel config is put in place,
the docker0 interface isn't re-IPed because docker sees the running
containers and won't update the previous config.
This patch kills all the running containers after docker is stopped.
We can't simply `docker stop` the running containers, as they respawn
before we've got a chance to stop the docker daemon, so we need to
use runc to do this after dockerd is stopped.
7 years ago
b999ee60aa
Fixing typo in kubelet cluster-dns and cluster-domain flags
7 years ago
85afd3ef14
Removing old sysv reference
7 years ago
1907030d89
issue raw yum command since we don't have facts in bootstrapping
7 years ago
71dabf9fb3
fix curl get calico status error ( error in tls version) : https://bugzilla.redhat.com/show_bug.cgi?id=1272504
7 years ago
efa2dff681
remove conditional
7 years ago
31a7b7d24e
default to kubedns and set nxdomain in kubedns deployment if that's the dns_mode
7 years ago
791ea89b88
Updated helm installation
Added full path for helm
7 years ago
0afbc19ffb
ensure the /etc/os-release is mounted read only
7 years ago
ac9290f985
add for rkt as well
7 years ago
a133ba1998
Updating calico to v2.1.4
7 years ago
5657738f7e
mount os-release to ensure the node's OS is what's seen in k8s api
7 years ago
883ba7aa90
Add support for different tags for netcheck containers
Replace 'netcheck_tag' with 'netcheck_version' and add additional
'netcheck_server_tag' and 'netcheck_agent_tag' config options to
provide ability to use different tags for server and agent
containers.
7 years ago
674b71b535
Ansible 2.3 support
- Fix when clauses in various places
- Update requirements.txt
- Fix README.md
Signed-off-by: Sergii Golovatiuk <sgolovatiuk@mirantis.com>
7 years ago
2638ab98ad
add MY_NODE_NAME variable into netchecker-agent environment
7 years ago
2bde9bea1c
Specify a dir and attach it to helm for HELM_HOME
7 years ago
041d4d666e
Install required selinux-python bindings in bootstrap
on centos. The bootstrap tty fixup needs it.
7 years ago
88b5065e7d
fix stray 'in' and break into multiple lines for clarity
7 years ago
b690008192
allow for correct aws default resolver
7 years ago
01dc6b2f0e
Add aws to default_resolver
When VPC is used, external DNS might not be available. This patch change
behavior to use metadata service instead of external DNS when
upstream_dns_servers is not specified.
Signed-off-by: Sergii Golovatiuk <sgolovatiuk@mirantis.com>
7 years ago
d8aa2d0a9e
Change DNS policy for kubernetes components
According to code apiserver, scheduler, controller-manager, proxy don't
use resolution of objects they created. It's not harmful to change
policy to have external resolver.
Signed-off-by: Sergii Golovatiuk <sgolovatiuk@mirantis.com>
7 years ago
e796cdbb27
Fix restart kube-controller ( #1242 )
kubernetesUnitPrefix was changed to k8s_* in 1.5. This patch reflects
this change in kargo
7 years ago
2d44582f88
Add tags to reset playbook and make iptables flush optional
Fixes #1229
7 years ago
b60a897265
Explicitly create cni bin dir
If this path doesnt exist, it will cause kubelet to fail to start when
using rkt
7 years ago
d68cfeed6e
Move namespace file to template directory
7 years ago
72d5db92a8
remove stray spaces in templating
7 years ago
3f302c8d47
ensure spacing on string of flags
7 years ago
04a769bb37
ensure spacing on string of flags
7 years ago
f9d4a1c1d8
update to safeguard against accidentally passing string instead of list
7 years ago
49be805001
allow admission control plug-ins to be easily customized
7 years ago
gdmelloatpoints
AtzeDeVries
Seungkyu Ahn
Spencer Smith
Gregory Storme
Brad Beam
Justin Hunthrop
zouyee
Josh Lothian
Matthew Mosesohn
xuhuilong
moss2k13
Aleksandr Didenko
Sergii Golovatiuk
Aleksey Kasatkin
FengyunPan
Greg Althaus
Sergii Golovatiuk
Hans Kristian Flaatten
gbolo